- slime-devel - mailman3.common-lisp.net

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 26 Aug '06

26 Aug '06

Index: slime/ChangeLog diff -u slime/ChangeLog:1.933 slime/ChangeLog:1.934 --- slime/ChangeLog:1.933 Thu Aug 24 08:15:59 2006 +++ slime/ChangeLog Fri Aug 25 11:44:45 2006 @@ -1,3 +1,9 @@ +2006-08-25 Kai Kaminski <kai.kaminski(a)gmx.de> + + * swank.lisp (lookup-presented-object): Fix for OpenMCL 1.0 + [ppc32], which requires that the :NO-ERROR clause is last in + HANDLER-CASE. + 2006-08-24 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> * slime.el (slime-ensure-presentation-overlay): Provide a

1 0

[slime-devel] Workaround for possible HANDLER-CASE bug in OpenMCL
by Kai Kaminski 25 Aug '06

25 Aug '06

Hi, I just tried to use the latest Slime with OpenMCL 1.0 (ppc32) and it choked on a HANDLER-CASE form in SWANK.LISP, because the :NO-ERROR clause is not the last. According to the Hyperspec that is legal, but OpenMCL doesn't like it anyway. Swapping the two clauses in the HANDLER-CASE form solves the problem (at least for me). I've attached a patch. Thanks, Kai

2 1

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 25 Aug '06

25 Aug '06

Index: slime/ChangeLog diff -u slime/ChangeLog:1.932 slime/ChangeLog:1.933 --- slime/ChangeLog:1.932 Tue Aug 22 05:34:12 2006 +++ slime/ChangeLog Thu Aug 24 08:15:59 2006 @@ -1,3 +1,28 @@ +2006-08-24 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> + + * slime.el (slime-ensure-presentation-overlay): Provide a + help-echo for presentations, showing the mouse bindings. + (slime-presentation-around-click): New function. + (slime-copy-or-inspect-presentation-at-mouse) + (slime-inspect-presentation-at-mouse) + (slime-copy-presentation-at-mouse) + (slime-describe-presentation-at-mouse) + (slime-pretty-print-presentation-at-mouse): New commands. + (slime-copy-presentation-at-point): Removed (misnomer). + (slime-presentation-map): Bind mouse-2 to + slime-copy-or-inspect-presentation-at-mouse, so the right thing is + done in REPL buffers and in Inspector and Debugger buffers. + (slime-menu-choices-for-presentation): Use the new commands here + instead of inline lambdas. + (sldb-inspect-in-frame): Use slime-read-object here, so if point + is in a presentation in the debugger buffer, inspect it + immediately just like slime-inspect does. + (slime-inspect-presented-object): Removed. + (slime-inspect): Don't expect that "swank:init-inspector" is + already part of the form. Accept an optional arg "no-reset". + (slime-read-object): Don't add "swank:init-inspector" to the read + form; slime-inspect now adds it. + 2006-08-21 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> Make the values of local variables in debugger frames and values

1 0

[slime-devel] slime and SBCL on FC2 (2.6.6-1.435smp) w/ XEmacs fails to startup right
by Sundar Narasimhan 25 Aug '06

25 Aug '06

Hi, when I run my sbcl -- which runs fine from the command line w/ Slime (devhead) I get a.. "Process inferior-lisp hangup" "Lisp connection exited abnormally w/ code 256". How can I go about debugging this? I see this mysterious: "There are also problems with kernel versions 2.6.5 - 2.6.10." in the PROBLEMS file. Thoughts? -Sundar

2 2

[slime-devel] SBCL takes 100% of cpu
by Alan Pavicic 25 Aug '06

25 Aug '06

Hello, after I enter 'slime-disconnect' my sbcl goes on 100% of cpu. I am running sbcl 0.9.15 and two different versions of slime from gentoo tree: slime-2.0.0.20060805 and slime-cvs. That's happening on two different machines. I tried to strace what is going on but sbcl crashed with 'Argh! corrupted error depth, halting' Following lines are from my .sbclrc: (setf swank:*use-dedicated-output-stream* nil) (setf swank:*communication-style* :fd-handler) (swank:create-server :dont-close t) Any ideas? Thanks, Alan Pavicic

2 1

[slime-devel] (setq swank:*globally-redirect-io* t) no longer works?
by Lars Rune Nøstdal 23 Aug '06

23 Aug '06

Hello, Using latest CVS of SBCL and Slime with swank:*globally-redirect-io* set to T no longer directs output to *standard-output* to the Emacs-buffer *slime-repl sbcl* when printing in a thread; it gets printed in the Emacs-buffer *inferior-lisp* instead. This happens when I print to *standard-output* from a thread, like this: cl-user> (sb-thread:make-thread (lambda () (write-line "hello?"))) #<sb-thread:thread {B0FC379}> Now I know I can fix this by binding *standard-output* inside the lambda via a lexical variable pointing to the "original" *standard-output* "outside" the thread. I'm wondering if this change of behavior of swank:*globally-redirect-io* is intended or maybe caused by something else? I've found this feature useful when doing quick and nasty debugging - so I kind of miss it. :) -- Lars Rune Nøstdal http://lars.nostdal.org/

1 0

[slime-devel] Using SLIME with OpenMCL with Obj-C bridge
by Phil 23 Aug '06

23 Aug '06

I'm having a minor problem using SLIME with OpenMCL and it's Objective-C bridge. Specifically, I want the REPL and any other threads that get created by SLIME to process commands via OpenMCL to be wrapped by a with-autorelease-pool call (which is defined by the bridge) to prevent memory leakage when creating Obj-C objects. Where would be the appropriate places in swank-openmcl to accomplish this? Thanks, Phil

2 2

[slime-devel] Help, can't get slime working.
by Dustin Robert Kick 23 Aug '06

23 Aug '06

I'm running Mac OS X 10.4.7, and I've tried getting slime to work with the OS's emacs, xemacs and emacs21 from fink, as well as a Carbon Emacs package that was linked from the Apple downloads site. I've tried the slime from fink, slime from cvs, and downloaded tgz, and the slime that came bundled in the Carbon Emacs. I've tried using clisp, cmucl, sbcl, and openmcl, and nothing I've tried works. Below is my current .emacs file, which is set to use slime from the tgz file, and sbcl from fink. If anyone can offer any help with any particular slime on Mac OS X setup that I've tried, please let me know, and I can show specific error messages I've gotten, if they might help. (setq auto-mode-alist (append '(("\\.mm$" . objc-mode) ;;("\\.lisp$" . fi:common-lisp-mode) ("\\.lisp$" . lisp-mode)) auto-mode-alist)) (add-to-list 'load-path "/Users/smee/slime-2.0") (require 'slime) (setq inferior-lisp-program "/sw/bin/sbcl" lisp-indent-functioin 'common-lisp-indent-function slime-complete-symbol-function 'slime-fuzzy-complete-symbol slime-startup-animation t) ;;;(add-hook 'lisp-mode-hook (lambda () (slime-mode t))) (add-hook 'inferior-lisp-mode-hook (lambda () (inferior-slime-mode t))) ;; Optionally, specify the lisp program you are using. Default is "lisp" ;;;(slime-setup) (add-hook 'lisp-mode-hook (lambda () (require 'slime) (slime-mode t) (slime-setup))) (setq load-path (cons (expand-file-name "~/ilisp-20021222/") load- path)) (autoload 'openmcl "ilisp" "Inferior Open MCL" t) ;; See the file "ilisp-5.11/ilisp.emacs" for ;; examples of other customization settings ;; that affect ILISP. Some recent ;; versions of XEmacs can be configured to ;; use a ~/.xemac/init.el file ;; instead of ~/.emacs (custom-set-variables ;; custom-set-variables was added by Custom. ;; If you edit it by hand, you could mess it up, so be careful. ;; Your init file should contain only one such instance. ;; If there is more than one, they won't work right. ) (custom-set-faces ;; custom-set-faces was added by Custom. ;; If you edit it by hand, you could mess it up, so be careful. ;; Your init file should contain only one such instance. ;; If there is more than one, they won't work right. '(default ((t (:stipple nil :background "black" :foreground "white" :inverse-video nil :box nil :strike-through nil :overline nil :underline nil :slant normal :weight normal :height 120 :width normal :family "apple-monaco")))) '(cursor ((t (:background "darkgrey"))) t)) (setq make-backup-files nil) (setq inhibit-startup-message t) (load-file "/Volumes/BRUNOISE.JARDIN/TIGRE/emacs/test.el") (load-file "/Volumes/BRUNOISE.JARDIN/TIGRE/emacs/test_commands.el") (fset '{ (lambda (&optional arg) "Keyboard macro." (interactive "p") (kmacro-exec-ring-item (quote ("{}{" 0 "%d")) arg))) (global-set-key (kbd "C-{") '{) (fset '} (lambda (&optional arg) "Keyboard macro." (interactive "p") (kmacro-exec-ring-item (quote ([19 125 left right return] 0 "%d")) arg))) (global-set-key (kbd "C-}") '}) (fset '\[ (lambda (&optional arg) "Keyboard macro." (interactive "p") (kmacro-exec-ring-item (quote ([91 93 18 91 right] 0 "%d")) arg))) (global-set-key (kbd "M-[") '\[) (fset '\] "\C-s]\C-m") (global-set-key (kbd "M-]") '\]) (fset '\) "\C-s)") (global-set-key (kbd "C-)") '\)) (fset '\( (lambda (&optional arg) "Keyboard macro." (interactive "p") (kmacro-exec-ring-item (quote ([40 41 18 40 right] 0 "%d")) arg))) (global-set-key (kbd "C-(") '\() Dustin Kick

4 3

[slime-devel] Daily ChangeLog diff
by mbaringer＠common-lisp.net 23 Aug '06

23 Aug '06

Index: slime/ChangeLog diff -u slime/ChangeLog:1.931 slime/ChangeLog:1.932 --- slime/ChangeLog:1.931 Sun Aug 20 05:37:13 2006 +++ slime/ChangeLog Tue Aug 22 05:34:12 2006 @@ -1,3 +1,32 @@ +2006-08-21 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> + + Make the values of local variables in debugger frames and values + of parts in the inspector accessible as presentations. In + particular, this allows to copy #<unreadable> values to the REPL + for further investigation. It also provides a context menu for + the values, offering to inspect, pretty-print, and describe them. + + Note that the presentations are only valid as long as the + corresponding Inspector or Debugger buffer is open. + + * swank.lisp (lookup-presented-object): Handle presentation ids + (:frame-var frame index), (:inspected-part part-index). + (init-inspector): New optional argument, reset. + + * slime.el (slime-inspector-insert-ispec): Mark up all values of + inspected parts as presentations. + (sldb-insert-locals): Mark up the values of local variables as + presentations. + (slime-remove-presentation-properties): Fix for read-only buffers. + (slime-copy-presentation-at-point): Make it work when the current + buffer is not the REPL buffer. + (slime-menu-choices-for-presentation): Describe into a separate + buffer, not the REPL. New menu item, pretty-print. + (slime-presentation-expression): Handle presentation ids that are + not numbers. + (slime-inspect-presented-object): Don't reset the inspector if + already in the inspector buffer. + 2006-08-20 Matthias Koeppe <mkoeppe(a)mail.math.uni-magdeburg.de> * swank.lisp (*nil-surrogate*): New. @@ -146,7 +175,6 @@ * swank.lisp (keywords-of-operator): New support function for writing user-defined `extra-keywords' methods. ->>>>>>> 1.922 2006-07-11 Helmut Eller <heller(a)common-lisp.net> * swank-allegro.lisp (make-weak-key-hash-table): Use ACL's weak

1 0

[slime-devel] Documentation for classes and slots
by Jonathon McKitrick 22 Aug '06

22 Aug '06

Is there a way to view the documentation strings for a class? I'd like to be able to hit C-c C-d d on a class name and see the doc strings for the class and the slots. Is this a feature I've just missed? Jonathon McKitrick -- My other computer is your Windows box.

2 1